删除 'Bug476-ZhangWeiHao-BaoYuelin'

app/difficulty.py

@@ -8,7 +8,6 @@
 import pickle
 import math
 from wordfreqCMD import remove_punctuation, freq, sort_in_descending_order, sort_in_ascending_order
-import snowballstemmer
 
 
 def load_record(pickle_fname):
@@ -19,12 +18,6 @@ def load_record(pickle_fname):
 
 
 def difficulty_level_from_frequency(word, d):
-    """
-    根据单词的频率进行难度的评级
-    :param word:
-    :param d:
-    :return:
-    """
     level = 1
     if not word in d:
         return level
@@ -37,44 +30,26 @@ def difficulty_level_from_frequency(word, d):
     return level
 
 
-def get_difficulty_level_for_words_and_tests(dic):
-    """
-    对原本的单词库中的单词进行难度评级
-    :param dic: 存储了单词库pickle文件中的单词的字典
-    :return:
-    """
-    d = {}
-    L = list(dic.keys())  # in dic, we have test types (e.g., CET4,CET6,BBC) for each word
-
-    for k in L:
-        if 'CET4' in dic[k]:
-            d[k] = 4  # CET4 word has level 4
-        elif 'CET6' in dic[k]:
-            d[k] = 6
-        elif 'BBC' in dic[k]:
-            d[k] = 8
-        print(k, d[k])
-
-    return d  # {'apple': 4, ...}
-
 def get_difficulty_level(d1, d2):
-    """
+    d = {}
-    d2 来自于词库的27000个已标记单词
+    L = list(d1.keys())  # in d1, we have freuqence for each word
-    d1 你个老六不会的词
+    L2 = list(d2.keys()) # in d2, we have test types (e.g., CET4,CET6,BBC) for each word
-    """
+    L.extend(L2)
-    d2 = get_difficulty_level_for_words_and_tests(d2)  # 根据标记评级，仅适用于词库中的词
+    L3 = list(set(L)) # L3 contains all words
-    stem = snowballstemmer.stemmer('english')
+    for k in L3:
+        if k in d2:
+            if 'CET4' in d2[k]:
+                d[k] = 4 # CET4 word has level 4
+            elif 'CET6' in d2[k]:
+                d[k] = 6
+            elif 'BBC' in d2[k]:
+                d[k] = 8
+                if k in d1: # BBC could contain easy words that are not in CET4 or CET6.  So 4 is not reasonable.  Recompute difficulty level.
+                    d[k] = min(difficulty_level_from_frequency(k, d1), d[k])
+        elif k in d1:
+            d[k] = difficulty_level_from_frequency(k, d1)
 
-    for k in d1:        # k是用户不会的词
+    return d
-        for l in d2:     # l是已经完成评级的词库的词
-            if k == l:  # k == l，这个用户也不会的词刚好以原型的形式出现在词库中，因为词库已经评过难度了，所以啥也不用干
-                break
-            elif stem.stemWord(k) in l:     # 这个词的词根与词库中的某个词一样，我们认为是同一难度的词
-                d1[k] = d2[l]
-            else:   # 这个词不在词库中，按频率来评定难度
-                d2[k] = difficulty_level_from_frequency(k, d1)
-
-    return d2